Skip to Main Content
The video applications have their own characteristics, and some real time applications need high QoS. But the Internet is very complex and sometime the supplies can not be provided. Recently, many researches have been found that much traffic is self-similar over several time scales. The bandwidth of network is tempi present. We use Kalman filter to filter the sending rate and to predict the rate while the ack-packets lost. In this paper, we propose a novel video transmission control algorithm, AFEC-Advanced FEC, which is integrated with FEC coding technology and Kalman filter theory. The algorithm modifies the rate with the adapted Kalman filter and the number of sending FEC packets is adapted by another mechanism. This algorithm can solve the collapse caused by the mass ACK packets and the rate fluctuation caused by the loss of the ACK packets. This algorithm also solves the problem of the elementary-layer-loss in transmission and keeps the continuity of the video transmission through the Internet. At last, we give the simulation results, which indicate the algorithm can provide good video services in loss-ratio-high channel.